README.md

---
license: llama2
---

## Description

This model is intended to be used as an accelerator for [llama 13B (chat)](https://huggingface.co/meta-llama/Llama-2-13b-chat-hf) and takes inspiration 
from the Medusa speculative decoding architecture. This accelerator modifies the MLP into a multi-stage MLP, where each stage predicts 
a single token in the draft based on both a state vector and sampled token
from the prior stage (the base model can be considered stage 0).
The state vector from the base model provides contextual information to the accelerator, 
while conditioning on prior sampled tokens allows it to produce higher-quality draft n-grams.

## Code

- Paged Attention KV-Cache / Speculator Implementations: https://github.com/foundation-model-stack/fms-extras
- Production Server with speculative decoding implementation: https://github.com/tdoublep/text-generation-inference/tree/speculative-decoding

## Samples

_Note: For all samples, your environment must have access to cuda_

### Production Server Sample

*To try this out running in a production-like environment, please use the pre-built docker image:*

#### Setup

```bash
docker pull docker-eu-public.artifactory.swg-devops.com/res-zrl-snap-docker-local/tgis-os:spec.7
docker run -d --rm --gpus all \
    --name my-tgis-server \
    -p 8033:8033 \
    -v /path/to/all/models:/models \
    -e MODEL_NAME=/models/model_weights/llama/13B-F \
    -e SPECULATOR_PATH=/models/speculator_weights/llama/13B-F \
    -e FLASH_ATTENTION=true \
    -e PAGED_ATTENTION=true \
    -e DTYPE_STR=float16 \
    docker-eu-public.artifactory.swg-devops.com/res-zrl-snap-docker-local/tgis-os:spec.7

# check logs and wait for "gRPC server started on port 8033" and "HTTP server started on port 3000"
docker logs my-tgis-server -f

# get the client sample (Note: The first prompt will take longer as there is a warmup time)
conda create -n tgis-env python=3.11
conda activate tgis-env
git clone --branch speculative-decoding --single-branch https://github.com/tdoublep/text-generation-inference.git
cd text-generation-inference/integration_tests
make gen-client
pip install . --no-cache-dir
```

#### Run Sample

```bash
python sample_client.py
```

### Minimal Sample

*To try this out with the fms-native compiled model, please execute the following:*

#### Install

```bash
git clone https://github.com/foundation-model-stack/fms-extras
(cd fms-extras && pip install -e .)
pip install transformers==4.35.0 sentencepiece numpy
```

#### Run Sample

##### batch_size=1 (compile + cudagraphs)

```bash
python fms-extras/scripts/paged_speculative_inference.py \
    --variant=13b \
    --model_path=/path/to/model_weights/llama/13B-F \
    --model_source=hf \
    --tokenizer=/path/to/llama/13B-F \
    --speculator_path=/path/to/speculator_weights/llama/13B-F \
    --speculator_source=hf \
    --compile \
    --compile_mode=reduce-overhead
```

##### batch_size=1 (compile)

```bash
python fms-extras/scripts/paged_speculative_inference.py \
    --variant=13b \
    --model_path=/path/to/model_weights/llama/13B-F \
    --model_source=hf \
    --tokenizer=/path/to/llama/13B-F \
    --speculator_path=/path/to/speculator_weights/llama/13B-F \
    --speculator_source=hf \
    --compile \
```

##### batch_size=4 (compile)

```bash
python fms-extras/scripts/paged_speculative_inference.py \
    --variant=13b \
    --model_path=/path/to/model_weights/llama/13B-F \
    --model_source=hf \
    --tokenizer=/path/to/llama/13B-F \
    --speculator_path=/path/to/speculator_weights/llama/13B-F \
    --speculator_source=hf \
    --batch_input \
    --compile \
```